Skoglunda and Jakobssona postulated in a paper  published in 2011 that the apparent inflow of archaic genes into modern humans in two separate episodes (one involving Neandertals and the other Denisovans) is not totally correct, at least regarding Neanderthals.
Their computer simulations suggest that an archaic admixture event (not out of Africa, but in Africa) would result in the current pattern of higher similarity with Neanderthals the further you go from Africa.
This would be caused by genetic drift and ascertainment bias, and the outcome is "artificial differences between populations that have exactly the same admixture history." 
However the Denisovan genes in Austronesians and South East Asians could not be explained away by the same cause and the "results suggest admixture between Denisovans or a Denisova-related population and the ancestors of East Asians..." 
Ascertainment bias is a systematic distortion in measuring the true frequency of a phenomenon due to the way in which the data are collected: the sampling may cause some members of the population to be less likely to be included in the sample than others. This distorts the randomness of the population, which is not equally balanced or objectively represented.
Their study defines two principal components (PCs) using Denisova, Neandertal, chimpanzee and modern human data.
They obtained a PC1 component which describes general genetic similarity to archaic humans (genomes of both Neandertals and Denisovans - it separates the archaic humans from chimps) and a PC2 which contrasts genetic similarity between Denisovans and Neandertals (it separates Denisovans from Neandertals).
- That moden humans can be grouped into three clusters: Africans, Oceanians and the other non-Africans. See Figure 1B below
- That PC1 and PC2 were correlated with geography in America and Eurasia, they find this incompatible with existing admixture theories since they expect homogeneous admixture of archaic ancestry in all non-Africans.
- The "East Asian [...] and Native American [...] populations were found to be more similar to archaic hominins compared with European and Central/South Asian populations"
- Also the further the populations were from our African homeland, the greater the "archaic ancestry signal". Seen in Fig. 1C
Figures taken from .
Notice in 1B how Native Americans are closer to Neanderthals and further from Denisovans (to which the Oceanians are closer).
Fig. 1C, also shows that both these groups have a greater "signal" of archaic ancestry. (Click on image above to enlarge)
Fig. 1D, shows a high Denisovan frequency in Oceania and South East Asia as well as the expected high Neandertal frequency in their former homeland (Europe and Middle East, North Africa), but unexpectedly high in Central America, Mexico and northern South America.
They suggest that these Neandertal genes came to America after its discovery by Europeans with the influx of settlers from that region ("individuals that were skewed to having European ancestry" ), and they add that the "remaining pattern of increased signs of archaic ancestry in American populations more distant from Africa is in line with the joint effect of ascertainment bias and genetic drift" .
It could also be possible that Neandertals migrated to America and settled there in Mesoamerica, admixing later with modern humans who migrated to America later through Beringia.
Surprisingly the authors state that there is an "apparent absence of Denisova ancestry in Native Americans" is probably due to the bias that makes Neanderthal genome higher in America due to genetic drift and the ascertainment bias.
As far as I can see, there is a high Neandertal frequency but also a high Denisovan one too.
Taking a look at Fig 1E, which replicates the same data as Fig 1D in Europe, SE Asia and Oceania, it clearly shows a high Denisovan frequency in Northwestern South America (Colombia, Ecuador)
Could it be due to an influx of SE Asian navigators' genes in the region (a junk sailing across the Pacific from China to Peru? or does it reflect an ancient migration of Denisovans into America long before modern humans ventured into the region to admix there, in America, with this relict Denisovan population?
If the latter was the case, then it is not a statistical artifact (bias in sampling), but a "real" admixture of genes. The Lack of Denisovan genes in Northern Asia is due to the fact that these genes did not arrive in America via Beringia carried by modern humans, they came long before, inside the bodies of the migrating Denisovans.
The authors however elaborate further by saying that if Americans lack Denisovan genes then their ancestors left Asia before admixture and, tainted with what I call the "recent-peopling-of-America-bias" they place this event some 14- 30 kya.
The maps clearly show an "island" between Mexico and Ecuador where there is a higher proportion of both Neanderthal and Denisovan genes in America. Is it due to a statistical bias? or is it faint proof of the ancient peopling of America by those archaic humans?
Our next post will deal with some odd skulls that were discovered in that region of South America, in Ecuador, and a possible link to H. erectus.
 Pontus Skoglunda, and Mattias Jakobssona, (2011) Archaic human ancestry in East Asia. www.pnas.org/cgi/doi/10.1073/pnas.1108181108
Patagonian Monsters - Cryptozoology, Myths & legends in Patagonia Copyright 2009-2013 by Austin Whittall ©